Short-term load prediction and restoration method based on historical load data

The invention discloses a short-term load prediction and restoration method based on historical load data, and the method comprises the steps: carrying out the data preprocessing of a historical load data set, and obtaining a preprocessed historical load data set; performing feature extraction processing on the preprocessed historical load data set through a depth auto-encoder to obtain a reconstructed data set; performing load prediction on the reconstructed data set through a Stacking integration model to obtain a prediction value; and repairing the historical load data set according to the predicted value to obtain a repaired historical load data set. According to the invention, the detection precision of the abnormal load data can be improved by accurately extracting the potential features of the historical load data. The short-term load prediction and restoration method based on the historical load data can be widely applied to the technical field of power load data mining. 本发明公开了一种基于历史负荷数据的短期负荷预测与修复方法，该